Hyperspectral Image Classification based on Dimensionality Reduction and Swarm Optimization

Murinto, Murinto and ismi, Dewi Pramudia and Iman HU, Erik (2019) Hyperspectral Image Classification based on Dimensionality Reduction and Swarm Optimization. [Artikel Dosen]

Text HASIL CEK_Murinto_Hyperspectral Image Classification based on Dimensionality Reduction and Swarm Optimization Approach.pdf Download (2MB)

Abstract

Hyperspectral images have high dimensions, making it
difficult to determine accurate and efficient image
segmentation algorithms. Dimension reduction data is done to
overcome these problems. In this paper we use Discriminant
independent component analysis (DICA). The accuracy and
efficiency of the segmentation algorithm used will affect final
results of image classification. In this paper a new method of
multilevel thresholding is introduced for segmentation of
hyperspectral images. A method of swarm optimization
approach, namely Darwinian Particle Swarm Optimization
(DPSO) is used to find n-1 optimal m-level threshold on a
given image. A new classification image approach based on
Darwinian particle swarm optimization (DPSO) and support
vector machine (SVM) is used in this paper. The method
introduced in this paper is compared to existing approach. The
results showed that the proposed method was better than the
standard SVM in terms of classification accuracy namely
average accuracy (AA), overall accuracy (OA and Kappa
index (K).

Item Type:	Artikel Dosen
Keyword:	Evolutionary algorithms, Segmentation, Classification
Subjects:	Q Science > Q Science (General)
Divisi / Prodi:	Faculty of Industrial Technology (Fakultas Teknologi Industri) > FTI Doc
Depositing User:	murinto murinto
Date Deposited:	18 Apr 2023 09:44
Last Modified:	23 Sep 2023 06:23
URI:	http://eprints.uad.ac.id/id/eprint/42978

Actions (login required)

View Item